Share your thoughts, 1 month free Claude Pro on usSee more

Table Reasoning Accuracy on TableBench

63.48Accuracy

SERO

Updated 5d ago

Evaluation Results

Method	Links
SERO 2026.05		63.48	-
Static Role Orchestration 2026.05		61.6	-
Static DAG MAS 2026.05		60.68	-
Random Role Evolution 2026.05		60.64	-
CoT 2026.05		59	-
SC@3 2026.05		58.5	-
Workflow 2026.05		57.91	-
14B Only 2026.05		54.85	42.4
SERO 2026.05		54.06	-
14B Only 2026.05		53.65	40.5
Static Role Orchestration 2026.05		52.43	-
Static DAG MAS 2026.05		52.3	-
EcoTab 2026.05		51.77	28.1
Random Role Evolution 2026.05		51.76	-
GlimpRouter 2026.05		50.52	30.1
EcoTab 2026.05		50.41	27.4
STEER 2026.05		50.33	29.4
SpecReason 2026.05		50.02	29.9
SpecCoT 2026.05		49.87	29.7
GlimpRouter 2026.05		49.54	28.1
SERO 2026.05		49.37	-
STEER 2026.05		49.22	29.1
RSD 2026.05		49.21	30.4
Workflow 2026.05		49.08	-
SpecReason 2026.05		48.89	28.4
RSD 2026.05		48.81	28.6
SpecCoT 2026.05		48.74	28.3
Random Role Evolution 2026.05		48.32	-
SC@3 2026.05		47.99	-
Static Role Orchestration 2026.05		47.95	-
CoT 2026.05		47.24	-
Random 2026.05		46.12	36.7
Workflow 2026.05		44.43	-
Static DAG MAS 2026.05		44.05	-
Random 2026.05		44.03	36.4
1.7B Only 2026.05		42.55	6.12
1.7B Only 2026.05		42.55	6.12
CoT 2026.05		29.44	-
SC@3 2026.05		29.27	-